Miguel-Angel Sicilia

Integrating reasoning and clinical archetypes using OWL ontologies and SWRL rules

Semantic interoperability is essential to facilitate the computerized support for alerts, workflow management and evidence-based healthcare across heterogeneous electronic health record (EHR) systems. Clinical archetypes, which are formal definitions of specific clinical concepts defined as specializations of a generic reference (information) model, provide a mechanism to express data structures in a shared and interoperable way. However, currently available archetype languages do not provide direct support for mapping to formal ontologies and then exploiting reasoning on clinical knowledge, which are key ingredients of full semantic interoperability, as stated in the SemanticHEALTH report [1]. This paper reports on an approach to translate definitions expressed in the openEHR Archetype Definition Language (ADL) to a formal representation expressed using the Ontology Web Language (OWL). The formal representations are then integrated with rules expressed with Semantic Web Rule Language (SWRL) expressions, providing an approach to apply the SWRL rules to concrete instances of clinical data. Sharing the knowledge expressed in the form of rules is consistent with the philosophy of open sharing, encouraged by archetypes. Our approach also allows the reuse of formal knowledge, expressed through ontologies, and extends reuse to propositions of declarative knowledge, such as those encoded in clinical guidelines. This paper describes the ADL-to-OWL translation approach, describes the techniques to map archetypes to formal ontologies, and demonstrates how rules can be applied to the resulting representation. We provide examples taken from a patient safety alerting system to illustrate our approach.

The semantic learning organization

Purpose – The aim of this paper is introducing the concept of a “semantic learning organization” (SLO) as an extension of the concept of “learning organization” in the technological domain.Design/methodology/approach – The paper takes existing definitions and conceptualizations of both learning organizations and Semantic Web technology to develop the new concept.Findings – The main points in which Semantic Web technology can be applied to learning in organizations are identified, and ontological accounts of organizational earning behaviour are pointed out as the main open question to develop the concept of a SLO.Originality/value – The paper provides a new conceptual framework for Semantic Web applications in organizational learning, which can be used as a roadmap for further research.

Integrating descriptions of knowledge management learning activities into large ontological structures: a case study

Ontologies have been recognized as a fundamental infrastructure for advanced approaches to Knowledge Management (KM) automation, and the conceptual foundations for them have been discussed in some previous reports. Nonetheless, such conceptual structures should be properly integrated into existing ontological bases, for the practical purpose of providing the required support for the development of intelligent applications. Such applications should ideally integrate KM concepts into a framework of commonsense knowledge with clear computational semantics. In this paper, such an integration work is illustrated through a concrete case study, using the large OpenCyc knowledge base. Concretely, the main elements of the Holsapple and Joshi KM ontology and some existing work on e-learning ontologies are explicitly linked to OpenCyc definitions, providing a framework for the development of functionalities that use the built-in reasoning services of OpenCyc in KM activities. The integration can be used as the point of departure for the engineering of KM-oriented systems that account for a shared understanding of the discipline and rely on public semantics provided by one of the largest open knowledge bases available.

Metadata, semantics, and ontology: providing meaning to information resources

Metadata research has emerged as a new discipline in the last years, and is focused on the provision of semantic descriptions of a diverse kind to digital resources, web resources being the most frequent target. Such associated descriptions are supposed to serve as a foundation for advanced, improved services in several application areas, including search and location, personalisation, and automated delivery of information. In consequence, metadata research focuses both on the development of metadata description languages of a general purpose or specialised kind and also on the practicalities of metadata creation, dissemination, assessment, maintenance, and use for diverse scenarios and usage contexts. Ontology has emerged recently as a knowledge representation infrastructure for the provision of shared semantics to metadata, which essentially forms the basis of the vision of the Semantic Web. The combination of metadata description techniques and ontology engineering defines a new landscape for information engineering with specific challenges and promising applications, which requires a truly multi-disciplinary approach. This paper is intended to provide some basic insights for the endeavour of engineering systems based on metadata, semantics, and ontologies, and to foster the interaction of researchers with different backgrounds coming from diverse disciplines.

Evaluating collaborative filtering recommendations inside large learning object repositories

Collaborative filtering (CF) algorithms are techniques used by recommender systems to predict the utility of items for users based on the similarity among their preferences and the preferences of other users. The enormous growth of learning objects on the internet and the availability of preferences of usage by the community of users in the existing learning object repositories (LORs) have opened the possibility of testing the efficiency of CF algorithms on recommending learning materials to the users of these communities. In this paper we evaluated recommendations of learning resources generated by different well known memory-based CF algorithms using two databases (with implicit and explicit ratings) gathered from the popular MERLOT repository. We have also contrasted the results of the generated recommendations with several existing endorsement mechanisms of the repository to explore possible relations among them. Finally, the recommendations generated by the different algorithms were compared in order to evaluate whether or not they were overlapping. The results found here can be used as a starting point for future studies that account for the specific context of learning object repositories and the different aspects of preference in learning resource selection.

Complete metadata records in learning object repositories: some evidence and requirements

Completeness of e-learning objects metadata records becomes a key requirement for learning object repositories, since these repositories are called to play a central role in automated approaches to e-learning. Nonetheless, metadata creation is a time-consuming and laborious process. These two factors may eventually result in incomplete and poorly structured metadata. In this paper, the completeness of learning object metadata of samples obtained from the MERLOT and CAREO repositories is analysed from that viewpoint, using the IEEE LOM standard as a reference framework. The paper concludes with a proposal for the specification of completeness levels as compliancy requirements for learning-related services or processes.

Learning processes and processing learning: from organizational needs to learning designs

Please, cite this publication as: Naeve, A. & Sicilia, M. A. (2006). Learning: from Organizational Needs to Learning Designs. Proceedings of Adaptive Hypermedia. June, Dublin, Ireland. Retrieved June 30th, 2006, from http://dspace.learningnetworks.org

Applying an ontology approach to IT service management for business-IT integration

Best practice frameworks focused on the integration of business and IT, such as ITIL, help organizations create and share effective service management. However, guidelines and models are commonly specified using natural language or graphical representations, both lacking the computational semantics needed to enable their automated validation, simulation or execution. This paper presents an ontology approach, which can help service providers add semantics to their service management process models and detect semantic ambiguities, uncertainties and contradictions. The proposed ontology draws its knowledge from good practice guidance for ITSM, enabling the current business gap that exists in many IT service providers to be overcome. To do so, service management processes are formalized in terms of an ontology defined using OWL combined with SWRL and SQWRL, the latter two being used to specify constraints and infer new knowledge. Our ontology provides support for executable service models with computational semantics. SWRL rules associated with the ontology can be categorized into three groups: (1) Model consistency; (2) SLA breaches; and (3) Proactive actions. Such rules allow us to better manage actual service management processes which are delivered in line with business needs. Also, the resulting specifications can be shared, reused and interchanged by automated means using e-business frameworks such as ebXML.

Computing with competencies: Modelling organizational capacities

The notion of competency provides an observable account of concrete human capacities under specific work conditions. The fact that competencies are subject to concrete kinds of measurement entails that they are subject to some extent to comparison and even in some sense, calculus. Then, competency models and databases can be used to compute competency gaps, to aggregate competencies of individuals as part of groups, and to compare capacities. However, as of today there is not a commonly agreed model or ontology for competencies, and scattered reports use different models for computing with competencies. This paper addresses how computing with competencies can be approached from a general perspective, using a flexible and extensible ontological model that can be adapted to the particularities of concrete organizations. Then, the consideration of competencies as an organizational asset is approached from the perspective of particular issues as competency gap analysis, the definition of job positions and how learning technology can be linked with competency models. The framework presented provides a technology-based baseline for organizations dealing with competency models, enabling the management of the knowledge acquisition dynamics of employees as driven by concrete and measurable accounts of organizational needs.

On the Convergence of Formal Ontologies and Standardized E-Learning

Current efforts to standardize e-Learning resources are centered around the notion of learning object as a piece of content that can be reused in diverse educational contexts. Several specifications for the description of learning objects – converging in the LOM standard – have appeared in recent years, providing a common foundation for interoperability and shared semantics. At the same time, the Semantic Web vision has resulted in a number of technologies grounded in the availability of shared, consensual knowledge representations called ontologies. As it has been proposed by several authors, ontologies can be used to provide a richer, logicsbased framework for the expression of learning object metadata, resulting in the convergence of both streams of research towards a common objective. In this paper, we address the practicalities of the representation of LOM metadata instances into formal ontologies, discussing the main technical and organizational issues that must be addressed for an effective integration of both technologies, and sketching some illustrative examples using modern ontology languages and a large knowledge base. On the Convergence of Formal Ontologies and e-Learning 3/19 On the Convergence of Formal Ontologies and Standardized e-Learning The increasing interest in Web-enabled education and training (what is often referred to as “eLearning”) has fostered a growing interest in the definition of specifications and reference models for digital educational contents, in an effort to standardize them (Anido et al., 2002). The objectives of such standardization efforts include facilitating their interchange, their composition and ultimately, their mass customization (Martinez, 2001). At the same time, the vision of a “Semantic Web” (Berners-Lee, Hendler & Lassila, 2001) has resulted in a renewed interest in the provision of shared, consensual knowledge representations (Davis, Shrobe, and Szolovits, 1993) for the annotation of documents, or, in a more general sense, of knowledge assets. The fact that e-Learning and Semantic Web technologies somewhat intersect has been raised in several recent research papers – including (Sicilia & Garcia, 2003a; Hoermann et al., 2003 and Stojanovic, Staab, and Studer, 2001) – that focus on the convergence of two key concepts: learning objects and ontologies. On the one hand, learning objects are an approach to instructional design centered in the notion of reusability (Wiley, 2001), understood as the capability for a digital content element (a learning object) to be used in several different learning situations, possibly in combination with other contents that were not originally designed for the same context (Sicilia & Garcia, 2003b). Learning object metadata records are used to describe technical requirements, educational properties and other kinds of information about learning objects, all of them in a standardized format, thanks to the availability of the LOM standard (IEEE, 2002) and several other conforming specifications. On the other hand, ontologies are logics-based consensual knowledge representations that are advocated as a mean to annotate Web resources (or electronic resources in general) to provide them with semantic, machine-understandable meaning, thus becoming enablers for knowledge management tools and processes (Fensel, 2002a). In consequence, and provided that learning objects are a specific kind of digital resource (with an explicit instructional intention), ontology-based annotations are a candidate for expressing learning object metadata records as sketched in (Sicilia & Garcia, 2003a). Since ontology formalisms are based on specialized logics (Baader et al., 2003), carefully designed for expressiveness and computational efficiency, they could be used to provide a richer (Kabel, 2001), semantic-enabled computation framework for metadata-based e-Learning. In fact, the use of formal ontologies to express metadata not only preserves current principles and practicalities that are applied to standardized metadata (Duval, Hodgins, Sutton, & Weibel, 2002), but also provides a richer framework for its realization and its subsequent use by automated tools. Principles like “modularization” and “extensibility” are addressed by the use of open XML-based formats prepared for the Web (Fensel, 2002a), while the principle of “refinement” is formally defined by the logical interpretation of subsumption. The practicalities of building application profiles for particular usages can be realized by means of constructing specialized ontologies from more general ones, adding specialized terms that restrict On the Convergence of Formal Ontologies and e-Learning 4/19 cardinalities, value spaces or relationships among metadata entities. In addition, the satisfiability requirements of a given ontology can be used as a mean to assess the completeness of metadata records. But the integration of formal ontologies with the paradigm of learning objects poses several problems that have not been addressed yet. These problems can be roughly categorized in technical and organizational issues. Technical issues include the practicalities of expressing the structure, properties and prospective contexts of use of learning objects as description-logics expressions, and thus entail a notion of what a complete and consistent metadata record should be. Organizational issues are those that would eventually be caused by the adoption of ontologybased learning objects in organizations, as part of an integrated value process (Lytras, Pouloudi & Poulymenakou, 2002b). These issues include the need for specific user interfaces, the provision of intra-organization conceptualizations coherent with shared ones and new approaches for the assessment of the quality of metadata records. This paper is intended as an attempt to provide an initial framework for further research regarding the integration of ontology-based annotation in the practice of designing and describing learning objects, adhering to the LOM standard as the high-level reference model of common metadata elements. To do so, LOM metadata elements are examined from the viewpoint of expressing them in description logics, and surrounding practical issues that may interfere with the adoption of such approach are described. The issues described here can be consider an specific aspect of the more general issue of the convergence of e-Learning and Knowledge Management that has been analyzed elsewhere (Maurer and Sapper, 2001; Lytras, Pouloudi and Poulymenakou, 2002). The rest of this paper is structured as follows. The second section describes related work and the motivation for the integration of formal ontologies with learning objects. The third section examines some of the principal technical and organizational issues that arise when trying to use formal ontologies to express LOM metadata elements with an educational intention. Specific metadata examples using the OpenCyc knowledge base and the W3C-OWL 1 language are provided as illustrations in the fourth section. Finally, conclusions and future research directions are sketched. Formal Ontologies as a Language for Learning Object Metadata The use of logics-based languages for metadata description lies at the hearth of the vision of a Semantic Web. But using such languages requires the understanding of specific annotation semantics, or alternatively, the provision of tools that hide these complexities to the average user by means of a carefully devised user interface. In addition, the semantics of current learning object metadata specifications are provided mainly through natural language explanation, so that the use of logic-based approaches requires some kind of support to provide a shared machine1 http://www.w3.org/TR/2002/WD-owl-ref-20020729/ On the Convergence of Formal Ontologies and e-Learning 5/19 understandable form to annotations. In this section, we review previous work that explicitly address the use of formal ontologies for the purpose of describing learning objects, and then, the main benefits of such approach are summarized.